Induction motor simultaneous fault diagnosis based on Takagi-Sugeno models

Abstract

This article proposes a model-based approach to the concurrent diagnosis of stator and rotor faults in induction motors (IMs) using Takagi-Sugeno (TS) fuzzy models. Fault-free detection is essential to prevent unexpected downtime and economic loss in industrial applications. The study first develops a dynamic model of the IM in the synchronized reference frame with the rotor under healthy and faulty operations. Different fault conditions like stator inter-turn short circuits, defective rotor bars, and combination thereof are considered. A TS model for every case is developed based on the precise nonlinear model. Simulation outcomes prove the validity of the new models in simulating the dynamic response of the motor under faulty operating modes. The residual signals are used to compare the performance of the model in fault isolation. The proposed method offers a classification that inherently separates between fault types. Such a contribution presents an efficient real-time fault detection and predictive maintenance facility, which renders it suitable for hardware-in-the-loop application in intelligent drive systems.